76 (3): 377 – 394 11.12.2018 © Senckenberg Gesellschaft für Naturforschung, 2018. A comparative analysis of the ventral nerve cord of Lithobius forficatus (Lithobiomorpha): morphology, neuroanatomy, and individually identifiable neurons Vanessa Schendel, Matthes Kenning & Andy Sombke* University of Greifswald, Zoological Institute and Museum, Cytology and Evolutionary Biology, Soldmannstrasse 23, 17487 Greifswald, Germany; Vanessa Schendel [[email protected]]; Matthes Kenning [[email protected]]; Andy Sombke * [andy. [email protected]] — * Corresponding author Accepted 19.iv.2018. Published online at www.senckenberg.de/arthropod-systematics on 27.xi.2018. Editors in charge: Markus Koch & Klaus-Dieter Klass Abstract. In light of competing hypotheses on arthropod phylogeny, independent data are needed in addition to traditional morphology and modern molecular approaches. One promising approach involves comparisons of structure and development of the nervous system. In addition to arthropod brain and ventral nerve cord morphology and anatomy, individually identifiable neurons (IINs) provide new charac- ter sets for comparative neurophylogenetic analyses. However, very few species and transmitter systems have been investigated, and still fewer species of centipedes have been included in such analyses. In a multi-methodological approach, we analyze the ventral nerve cord of the centipede Lithobius forficatus using classical histology, X-ray micro-computed tomography and immunohistochemical experiments, combined with confocal laser-scanning microscopy to characterize walking leg ganglia and identify IINs using various neurotransmitters. In addition to the subesophageal ganglion, the ventral nerve cord of L. forficatus is composed of the forcipular ganglion, 15 well-separated walking leg ganglia, each associated with eight pairs of nerves, and the fused terminal ganglion. Within the medially fused hemiganglia, distinct neuropilar condensations are located in the ventral-most domain. Immunoreactive neurons of different transmitter systems (allato- statin, histamine, and FMRF-amide) display serially homologous patterns that may lay the foundation for comparison with other arthropod taxa. Moreover, a pair of histaminergic neurons may constitute a promising intra- as well as interspecific IIN candidate. Key words. Arthropoda, centipedes, neuroanatomy, histology, immunohistochemistry, microCT, allatostatin, FMRF-amide, histamine. 1. Introduction The central nervous system of arthropods is character- pod sistergroup relationship (e.g. WÄGELE & KÜCK 2014). ized by subdivision into brain and ventral nerve cord Several morphological characters initially proposed to (RICHTER et al. 2010; LOESEL et al. 2013), showing taxon- be apomorphic for Tetraconata were found to be present specific transformations and adaptations that are central in representatives of Myriapoda as well, giving further to comparative neuroanatomical and phylogenetic stud- support for the Mandibulata-hypothesis (MÜLLER et al. ies (neurophylogeny sensu HARZSCH 2006). In this con- 2003, 2007; SCHACHTNER et al. 2005; SOMBKE et al. 2011a; text, myriapods play a crucial role, as their phylogenetic SOMBKE et al. 2012; SOMBKE & HARZSCH 2015). position still is a matter of debate. A widely accepted Along these lines, neuromorphological and neuro- concept, which will be followed here, places the Myri- anatomical characters proved to be a suitable character apoda within the taxon Mandibulata as the sistergroup of complex as they are considered rather robust over evo- a hexapod-crustacean clade (i.e. Tetraconata or Pancrus- lutionary time scales, and have been used in cladistic tacea-hypothesis: REGIER et al. 2010; EDGECOMBE 2010; analyses resulting in well-resolved, yet still debated ROTA-STABELLI et al. 2011; MISOF et al. 2014), as opposed arthropod phylogenies (STRAUSFELD & ANDREW 2011; to the Tracheata-hypothesis, proposing a myriapod-hexa- WOLFF et al. 2017). Detailed investigations on the ner- ISSN 1863-7221 (print) | eISSN 1864-8312 (online) 377 Schendel et al.: Ventral nerve cord of Lithobius Fig. 1. Lithobius forficatus and dissected nervous system. A: Adult male of Lithobius forficatus, dorsal view. B: Dissected ventral nerve cord with walking leg ganglia 2 to 15 and terminal ganglion, dorsal view. The ganglia associated with the forcipular and first walking leg segments are omitted. — Scale bar: 1 mm. — Abbreviations: G: ganglion, TG: terminal ganglion. vous system of Myriapoda, and specifically Chilopoda Five major lineages of Chilopoda are established. (centipedes), however, are few when compared to the Scutigeromorpha are considered to be the most basal wealth of studies conducted in Hexapoda and Crustacea taxon and sistergroup to Pleurostigmophora. The latter (SOMBKE & ROSENBERG 2016). One promising field for comprises Lithobiomorpha and Phylactometria (Crate- comparative neuroanatomical analyses is the study of rostigmomorpha and Epimorpha), with Epimorpha be- individually identifiable neurons (IINs) in the arthropod ing composed of Scolopendromorpha and Geophilo- brain and ventral nerve cord (VNC). IINs can be identi- morpha (EDGECOMBE 2011). Recently, this view has been fied and homologized intra- and interspecifically based challenged by transcriptomic data that place Lithobio- on the specific criteria of position and morphology of morpha as the sistergroup of Epimorpha, repositioning somata and neurites, ontogenetic aspects, as well as Craterostigmomorpha to a more basal node (FERNÁNDEZ physiological and biochemical characteristics (KUTSCH et al. 2016). The centipede Lithobius forficatus (Lin- & BREIDBACH 1994). In particular, the neurotransmitter naeus, 1758) is one of the best-studied myriapod species serotonin (5HT) has been studied comprehensively in with respect to general anatomy and morphology and a comparative context (THOMPSON et al. 1994; HARZSCH other biological fields, such as endocrinology, develop- & WALOSZEK 2000; HARZSCH 2004; STEGNER et al. 2014; ment, and physiology (MINEllI 2011). The trunk of L. BRENNEIS & SCHOLTZ 2015; STEMME et al. 2017; SOMBKE forficatus is composed of the forcipular segment, 15 leg- & STEMME 2017). The first studies on lobsters, locusts bearing segments, three fused postpedal segments bear- and cockroaches pointed out striking morphological ing a pair of gonopods, and a telson (MINEllI & KOCH similarities of serotonergic neurons, inferring homolo- 2011). The VNC is composed of 18 mostly well-sepa- gies at the single cell level (BELTZ & KRAVITZ 1983; rated ganglia: the subesophageal ganglion, the forcipu- BISHOP & O’SHEA 1983; TAGHERT & GOODMAN 1984). lar ganglion, 15 leg-associated ganglia, and the (fused) Thus easily assessable and low in neuron number, IINs terminal ganglion. Each walking leg ganglion possesses constitute a suitable character complex in evolution- eight pairs of nerves innervating e.g. the legs and trunk ary discussions (KUTSCH & BREID BACH 1994). Recent musculature (RIllING 1960, 1968; SOMBKE & STEMME studies on centipedes revealed apomorphic characters 2017). Using histological techniques, morphology of the for this taxon, but also synapomorphic characters with VNC was investigated sporadically from the early 19th Tetraconata in soma positions and neurite trajectories, century on (summarized in HIlkEN et al. 2011; SOMBKE with implications for the reconstruction of the mandib- et al. 2011b), but apart from the contributions by HECK- ulate ground pattern (SOMBKE & STEMME 2017). How- MANN & KUTSCH (1995) and SOMBKE & STEMME (2017), ever, when considering the at least 50 known different the internal neuroanatomy remains poorly known. Con- neuropeptides and other classes of neuroactive sub- sequently, neuroanatomical data that include specific la- stances present in arthropods (e.g. biogenic amines like beling of single neurons, or labeling against neuroactive histamine) (e.g. HOMBERG 1994; NÄSSEL 2000; GÄDE & substances in the brain and the VNC are sparse (HECK- MARCO 2006; CHRISTIE et al. 2011; CHRISTIE 2015), this MANN & KUTSCH 1995; MELZER et al. 1996; PETYKÓ et serotonin-centric view leaves potentially informative al. 1996; HARZSCH 2004; SOMBKE et al. 2012; SOMBKE & characters unconsidered. STEMME 2017). 378 ARTHROPOD SYSTEMATICS & PHYLOGENY — 76 (3) 2018 In order to gain deeper insights into centipede neuro- dine solution (iodine resublimated in 99% ethanol; Carl anatomy, we investigate the VNC of Lithobius forficatus. Roth, X864.1) for 12 hours. Preparations were washed In this multi-methodological approach, we use histo- several times in pure ethanol and critical point dried us- logy and X-ray micro-computed tomography, as well as ing the automated dryer Leica EM CPD300 (Leica Mi- histochemical and immunohistochemical experiments crosystems). Finally, samples were fixed on insect pins combined with confocal laser-scanning microscopy, to with super glue. Scans were performed with a Zeiss document external morphology and neuroanatomy of Xradia MicroXCT-200 at 30 kV, 6 W and 1 second (10 × the walking leg ganglia. Also, we visualize individually magnification) or 3 seconds (20 × magnification) ex- identifiable neurons using a set of antibodies against two posure. Scan parameters were: (1) mid-trunk with 10 × neuropeptide families (allatostatin, FMRF-amide) and objective lens unit resulting in 2.14 µm pixel size, (2) the biogenic amine histamine. As comparable data that posterior-trunk with 10 × objective lens unit resulting can be used to infer primary homology hypotheses are in 1.99